Project Summary The standard for assessing the effectiveness of drugs to treat metastatic melanoma is the patient's response, but there is a pressing clinical need for a human surrogate model that could support prediction of drug efficacy, thereby saving the patient from trial and error treatments, and that would ultimately serve as a guide for the selection of patient-targeted drug therapies. Today, there is significant interest in the use of patient-derived xenografts (PDXs), in which a patient's tumor is implanted into an immune-deficient mouse, to create in the mouse a model of the patient's tumor. Unfortunately, this process is slow and expensive and is based upon an animal microenvironment rather than a human one. Microphysiological systems (MPS), which encompass organs-on-chips, tissue chips, and engineered organoids, can be constructed using human cells to create an in vitro microenvironment. The proposed research would build upon a strong collaboration at Vanderbilt University, the University of Pittsburgh, and the University of Wisconsin to develop powerful MPS to address the need for models of a patient's response to cancer therapy. This project will study how the tissue microenvironment affects the growth of metastatic melanoma cells and their response to drugs by using the Vanderbilt neurovascular unit tissue chip, the Pittsburgh liver-on-chip, and the Wisconsin engineered organoids for brain and liver, each of which includes multiple cell types. The research will focus on the final stage in the metastatic cascade ? the growth of tumor cells at sites distant from the primary tumor. This growth is governed by ?seed and soil? interaction between the tumor ?seed? and the tissue microenvironment ?soil.? Instead of using a mouse as the soil, patients' cancer cells will be planted into the soil provided by brain and liver MPS constructs derived from human induced pluripotent stem cells. The aims are 1) Implement a common set of human organ constructs (liver-on-chip, neurovascular unit, and engineered organoid from a single human stem cell source), 2) Demonstrate successful seeding of these human organ constructs with metastatic cutaneous melanoma or uveal melanoma cells derived from Vanderbilt and Pittsburgh patients, and 3) Compare the response to drugs by patients' cancer cells that have been seeded into the organs-on-chips and engineered organoids with the response to the same drugs by existing PDX lines. This project will provide guidance as to which in vitro human model might be more predictive of patient outcome when translated to the clinic, based in part upon the type of tumor, the nature of the patient sample, and the patient genotype. It will also test the hypothesis that the human MPS devices and models developed at Vanderbilt, Pittsburgh, and Wisconsin will provide a more realistic, in vitro, three-dimensional human microenvironment to study tumor metastasis than mouse PDXs. The final phase will be a proof-of-concept demonstration of precision medicine in which the microenvironment of the brain and liver could be from the patient's induced pluripotent stem cells.